Applying Natural Language Processing and TRIZ Evolutionary Trends to Patent Recommendations for Product Design

Traditional TRIZ theory provides methods and processes for systematic analysis on engineering problems, which can improve the efficiency of solving problems. However, effect problems is not necessarily guaranteed, depends user’s profession experience. Therefore, this study proposes a methodology to apply evolutionary benefits in 37 trend lines developed by researchers assist intelligently screening relevant patents applicable content product design. In such way, problem design quality may be improved more effectively. First, patent database used as training dataset, words sentences documents are analyzed through natural language processing obtain keywords that related benefits. Using word vectors trained Doc2vec, semantic similarity calculated relationship between text benefit. Secondly, goals development project make benefits, then recommendations provided. The proposed achieve purpose intelligent assistance enhance process problem-solving.